The main findings of this study are that self-administered test and questionnaires are possible to use for testing and classification of women with suspected PGP. In our earlier trial , where both pregnant women with and without pain and non-pregnant women without pain were assessed the results indicated that the self-administered tests had high sensitivity and specificity. Based on both our trials, the tests and concept seems to be usable in larger surveys. In addition, they can be used in perinatal care units as a ground for referral to physical therapy or other treatments for pregnant women with suspected PGP. This could also reduce the mistrust which may occur between midwives and pregnant women if vague symptoms are reported .
Among the tests for identification of posterior PGP, the highest percentage of agreement and sensitivity was seen for the self-administered P4 test as compared to the traditional P4. The result is in accordance with previous studies [12, 20]. A reason for the high agreement may be the standardisation of the test and simplicity of its performance. Likewise, this may explain why the MAT test showed the highest percentage of agreement among the tests for identifying anterior PGP.
The bridging test is another assessment for identifying posterior PGP that has been shown to have a high sensitivity and high percentage of agreement compared to tests performed by an examiner. In our previous study, the bridging test had a higher sensitivity than the self-administered P4 test when compared to the traditional P4 test . In the current study, the sensitivity of the self-administered and examiner performed P4 test was 80.6% and the bridging test 75.5%. This indicates that it may be an advantage to use at least these two self-administered tests for identification of PGP, as it has been reported that two to three positive pain provocation tests are required for a clinical classification . On the other hand it is important to limit the number of tests used while the test may trigger the pain. It seems like it is enough to use the P4 and bridging test to encompass the posterior pain.
The ASLR test was included in this evaluation because it is used as a functional test to determine the load transfer between legs and lumbar spine. . The results indicate though that it is less suitable as a self-administered test, as some women gave a score for difficulty in lifting the leg at home and less difficulty when the test was repeated at the clinic. A possible explanation for our findings could be the test’s grading system, where total concurrence is harder to fulfil. A further analysis was then performed where the results of the tests were dichotomized. The percentage of the agreement was then 78.9% between the tests indicating that if the test is used self-administered it is better to ask if the patient has difficulties to raise the leg or not than to grade the difficulty from 0–5. None of the women had a positive SLR during the visit, but 23 registered a positive self-administered SLR test. A possible explanation may be that the self-administered SLR test gave unspecific muscle pain, which the women interpreted as radiating pain to the foot. A positive nerve root pain is rare among pregnant women and, to avoid false positive self-administered tests, a better description is needed of how to interpret pain in the test.
The women in our study were included at a specialist clinic for lumbopelvic pain and our results might be generalised to women who seek care for their pain during pregnancy. Among these women, there are probably many with a high risk of persistent pain postpartum [21, 22], since women with severe pain and disability are more likely to seek care for their symptoms than women with mild complaints. Our results are promising for women who need to be identified early for treatment. Patients with verified pain can then be referred for further examination and treatment.
The tests evaluated in this article were chosen according to recommendations in guidelines and clinical trials concerning tests [3, 12, 17]. However, there are several other tests for PGP that were not included. It may be possible to use some of them in a self-administered way by the women, with or without adjustments. In addition, there is a need for self-administered tests for other structures close to the pelvis that can cause pain during pregnancy, such as the hip joint and groin.
There were significantly higher numbers of positive P4 and bridging tests during the visit compared to the self-administered tests (P = 0.036 and 0.001 respectively) and significantly lower numbers of positive modified Trendelenburg tests during the visit (anterior p < 0.001, posterior p < 0.016). The larger number of positive tests at the clinic may be explained by a more specific test procedure during the P4 and bridging tests. The interaction between examiner and subject can also be a reason while there is a risk that the patients try to communicate to the examiner that there definitely is a problem, making tests more positive. The explanation for the discrepancy concerning the modified Trendelenburg test may be that unspecific pain in the pelvic region can be misinterpreted by women to be symphyseal or PGP.
There are two other reasons that may explain the discrepancy between the tests. In this study, the self-administered tests and tests performed during the standardised examination were not performed on the same day. In our earlier study , the women performed both sets of tests at the clinic. PGP is reported to be more severe during evenings [9, 23] and the discrepancy between the two series of tests may be caused by that the women performed the self-administered test in the evening and the test at the clinic was performed in mornings or afternoons. However, as the repeated tests may overload the structures and trigger PGP, thus giving false positive results, it can be an advantage not to repeat the tests on the same day. Another explanation may be that the self-administered tests were performed according to written instructions and photos. In our earlier trial, verbal instructions were given and the women could ask for further instructions when they needed them. In an attempt to standardise the tests in the current evaluation, the women were instructed to perform the tests on the floor so that they were on a solid surface, rather than doing the tests in a soft bed. The same instructions were given at the clinic.
One limitation of this study is that women with pain of discogenic origin may not have been identified but such origin of lumbopelvic pain in pregnancy is rare . Another limitation is that this trial was undertaken to evaluate the tests in women who was referred to a specialist clinic because of suspected PGP and not a cohort of pregnant women. However, it is a first evaluation of the tests performed by women themselves in a natural setting, eg their homes. More evaluations are needed to explore the tests usability in pregnant women with and without lumbopelvic pain and functional limitations. The tests usefulness for classification of PGP postpartum also needs to be evaluated further.
Since it seems possible to identify women at risk for persistent PGP after pregnancy already early in the pregnancy [25, 26], the self-administered test could contribute to specific identification of PGP and thereby provide the basis for an early intervention.